Climate dynamics
An investigation into the impact of geometrical constraints on the large-scale circulation of the ocean and climate.  This work is based on a series of idealized GFD experiments with a numerical model of the coupled ocean-atmosphere-sea ice system. In this way, we avoid the complications arising from a realistic representation of geography/land-sea distributions and focus on the coupling of the two fluids in its most elemental form.
We use narrow land barriers set into a flat-bottomed ocean to turn on/off various components of the ocean circulation (gyre, circumpolar current, inter-hemispheric meridional overturning...)
This system is a great tool to address fundamental aspects of climate dynamics such as: Why is deep water formation in the Northern hemisphere confined to the Atlantic basin? Can a complex coupled climate system sustain multiple equilibria or is this only a property of low order model (such as Energy-Balanced Model) ?
A movie of the Aquaplanet in action here.
More details about this series of experiments here and here.
Research Interest
Mesoscale Eddies in the ocean
(top) Surface wind stress and (bottom) ocean eddy stress (color) and mean zonal current (black contours). The unresolved eddy processes (the eddy stress here) are estimated by bringing a coarse resolution model in close agreement  with observed climatologies through an adjoint optimization. The eddy stress thus obtained is large in the Southern Ocean and Western Boundary currents; the eddy stress radiates the input of momentum at the surface (due to winds) downward in to the deep ocean where it is balanced by the topographic form stress. More details here.
Dynamics of the Ocean Heat Transport
Meridional overturning circulation within temperature layers  (top, in Sv) in an ocean GCM and (bottom) the heatfunction for the same solution (in PW). The heatfunction shows the pathway of energy transport in the ocean interior. Note that the  cold/deep overturning circulations seen in the top panel are not associated with any significant heat transport. More details here.
Air-sea interactions in the midlatitudes
This is mainly work done as a Ph.D student with Claude Frankignoul at LODYC (Paris). I studied the influence of the ocean on the low-frequency atmospheric variability in the midlatitudes.
This was first done by building a simple analytical model of the coupled ocean-atmosphere system from which we could extract possible mechanisms of coupled air-sea interactions and their statistical signatures. For example, thermocline fluctuations can modulates the SST in such way that leads to their reinforcement by the surface wind stress. In such case, the SST and atmospheric variability increase at decadal timescales. The ocean dynamics can result in a small, but significant, atmospheric predictability a few years in advancce.
Transient atmospheric response to an interactive SST anomaly in the North Atlantic sector. The atmospheric response is initially baroclinic, as predicted by linear theory, and then evolves into an equivalent-barotropic one.
In a second part, I studied more precisely the dynamics of the atmospheric response to SST anomalies, and in particular its time evolution.
Aqua Drake Double-Drake Ridge